Marine Environmental Impact on CFAR Ship Detection as Measured by Wave Age in SAR Images
Round 1
Reviewer 1 Report
The manuscript discussed a ship target detection method on SAR images. Main techniques applied included constant false alarm rate (CFAR), sea clutter density estimation via generalized gamma distribution, and fitness evaluation by Kolmogorov-Smirnov distance. The main contribution is the proposed methodology tried to introduce the wind and wave information the existing CFAR target detection method. The following suggestions are provided for authors to considerate.
1. Overall, the idea of including wind and wave information to CFAR algorithm is interesting. But the methodology discussed in the manuscript has no substantiated the aforementioned motivation. The major discussion regarding this issue were found as follows:
“Yet, its behavior in relation to different ocean environments of distinct wave and wind regimes, as incorporated in the WA parameter, is still an unexplored topic [18]. Herein, parameters estimation for the GΓD is performed analytically by the approximate estimator given [17], which is based on the method of log-cumulants (MoLC) and shows to be more computationally efficient compared to traditional estimators.”
I could not figure out how the wind and wave information are explored through the above sentences, and similarly in other parts of the manuscript.
2. Furthermore, according to the simulation results shown in Figure 2, it seems that the conventional density estimation by generalized gamma distribution can differentiate different wave situations effectively. In other words, there is no need to consider the wind or wave information since their densities already showed significant difference.
3. The presentation of the manuscript can be improved further. For example, Page 8, “The strategy adopted to estimate de f values for different WA image classes”, Page 14, caption of Figure 6 "Nota: as legendas da Fig. 6a parece não estarem corretas ... and etc.
Author Response
We thank the reviewer for his contributions. Please, check the file enclosed with our answers and adjustments.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper is generally well organized and structured, especially the first half of the paper. Provided explanations are clear and easy to follow. However, there are few points in paper that need to be improved:
- One of the listed contributions is: "2. Enhancement of CFAR performance by means of a discrimination scheme based on ocean conditions." This scheme should be described in same way used in section 2.2.1. In this way, the scheme will be more pointed out in the paper and it will provide better and clearer insight to the readers. Also, currently it is not clear if the parameters' values used in discrimination scheme are constant or not (i.e. depend on sea conditions and other factors).
- Table 1 contents and text in the paper do not match completely. Based on the text, the largest distance should be for swell class and not for old wind sea class but in Table 1 the numerical values show opposite. Clarify this mismatch.
- Results in 3.4 should be also presented in the table not just through text. That will improve the presentation of the results. From the given results it is not completely clear what is the number of false alarms. Figure 6b shows oil platforms, and others are classified as dark ships, but is it correct detection or false detection of a vessel i.e. which are considered as false alarms. AIS detected vessels and not with CFAR - what are the sizes of these CFAR undetected vessels - this would be interesting info. Near-range, mid-range, far-range distance could be also included in results and discussion to check false alarms dependence on distance. Also, results in 3.4 should include other classes i.e. old wind sea class and swell. In that way, the reader will get full picture across the wa classes regarding the false alarms.
The paper is mostly written well, but there are some minor errors such as typos, sentences/words left in Portuguese language. Also, not all abbreviations have full name at their first appearance in the paper. One more proofreading should resolve these errors.
Author Response
We thank the reviewer for his contributions. Please, check the file enclosed with our answers and adjustments.
Author Response File: Author Response.pdf
Reviewer 3 Report
The proposed algorithm involves analyzing statistical properties of the radar data and performing multiple calculations, which can increase the overall complexity of the system. So how do you handle the complexity?
In environments with high clutter levels, the false alarm rate may increase or decrease affecting the algorithm's performance. Is there a accurate option to detect background clutter.
Discuss how the algorithm's performance will not degrade when confronted with very small or low radar cross-section (RCS) targets.
Successfully implementing the CFAR algorithm often necessitates meticulous tuning of parameters, as well as frequent adjustments in relation to varying operational scenarios and environments. Furthermore, to optimize performance levels, extensive training with relevant data might be required an undertaking that can prove rather daunting due its time-intensive nature.
Lack of comprehensive and up-to-date wave data can hinder the reliability of using wave age for ship detection and situational awareness. Wave conditions can vary spatially and temporally within a given area. These variations can impact the accuracy of ship detection based on wave age, as the parameter may not adequately represent the current wave conditions.
Does Cross-Validation does not help to evaluate the algorithm's performance on unseen data and provides a more reliable estimate of its effectiveness?
Evaluation and effectiveness of the proposed algorithms ,methods and criteria can be elaborated (CFAR performance) in detail if possible.
Author Response
We thank the reviewer for his contributions. Please, check the file enclosed with our answers and adjustments.
Author Response File: Author Response.pdf
Reviewer 4 Report
Comments to the Author
1. rewrite the abstract from the beginnings by focus on the "idea, develop points in this method presented by author"s" and compare the results of this manuscript with other papers as values)
2. The introduction section could be improved by clarifying the similarities and differences between the related work and the proposed method are not clearly described. It is recommended to add a separate subsection and clear description in this regards.
3. multi figures and information appears as ambiguous and not clear
4. add new section under title the hypothesis and limitations of the develop method present by author(s)
5. multi points in this manuscript need justification and prove by the author(S), after each table or figure give specific description of result submitted by it described in three into five lines
6. add new section discuss on it the results with details and explain on it the main advantages and disadvantages of their method based on the author(s) opinion
7. the main benefit of previous works are to compare your work with it from points(techniques, preprocessing stage, results or evaluation measures), while the author(s) in this manuscript not compare their works with the previous works from any points. therefore, must add table at the final the related works analysis the previous work from points "techniques used, preprocessing techniques, type of dataset used, evaluation measures, advantage and disadvantages of that technique )
8. Readership appeal is moderate and there is scope for improvement using more pictorial and graphical representations.
9. Background information of this work can be provided more systematically and comprehensively, i.e. logic of this paper should be further enhanced.
10.Results and illustrations need to be revisited.
11.The conclusion section should be enhanced while presenting the findings and quantitative results
Moderate English Language editing is required.
Author Response
We thank the reviewer for his contributions. Please, check the file enclosed with our answers and adjustments.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors have carefully responded my previous comments. However, my major concern, i.e., the densities actually already implied the wave age information, still remains. The authors claimed that the PDF estimation may be computationally costly, but this is actually a typical signal processing necessary to any radar system. And the wave age data obtained from other approaches are usually costly and sometimes unreliable especially for remote areas. I think the authors should provide sufficient evidence, either from theoretical analysis or empirical testing, to support their methodology.
Author Response
Dear reviewer,
please check the file enclosed with our answers and clarifications.
Kind regards
Author Response File: Author Response.pdf
Reviewer 2 Report
I am satisfied with the revised paper and have no additional remarks.
Author Response
Dear reviewer,
please check the file enclosed with our answers and clarifications.
Kind regards
Author Response File: Author Response.pdf
Reviewer 4 Report
The authors have significantly improved the manuscript based on my comments. Therefore, in my opinion, the manuscript can be accepted.
I have not found any issue with the English Language.
Author Response
Dear reviewer,
please check the file enclosed with our answers and clarifications.
Kind regards
Author Response File: Author Response.pdf
